Conventional communication networks for transmitting voice signals are usually based on Public Switched Telephone Network (PSTN) using a Time Division Multiplexing (TDM) technique. This kind of communication network gives priority to circuit switching, and voice services are transmitted and controlled by switches. The structure of a switch in a conventional TDM-based circuit switching network is shown in FIG. 1, wherein Signaling System 7 (SS7) is widely used in call controlling.
Along with a fast development of data services, conventional circuit switching networks bring much inconvenience to data service users, such as a high communication cost, a low net surfing speed, a long waiting time, a poor transmitting quality, and difficulties of implementing new services. On the other hand, along with the development of communication techniques, it becomes more and more possible that one data network is the bearer of all services. It is in such a situation that the NGN is put forward and has seen fast development in recent years. The NGN has gradually changed the conventional PSTN circuit switching based network to a packet switching based network. This kind of packet switching based network can carry all kinds of services of the previous PSTN network, and at the same time, offload a great deal of data transmission to an Asynchronous Transfer Mode (ATM)/Internet Protocol (IP) network to release the burden of the PSTN. Moreover, based on new properties of ATM/IP, many new services are provided and functions of many old services are enhanced. In this sense, the NGN is an outcome of combining the TDM based PSTN voice network and the ATM/IP based packet network, which makes it possible to transmit voice, video, data and other integrated services in one new network. Moreover, the NGN adopts a softswitch technique in which service control is separated from media service bearers, which greatly enhances the processing capability of network equipments so that more voice services can be processed and the network can be the bearer of more kinds of services.
International Telecommunication Union-Telecommunication Standardization Sector (ITU-T) gave the following definitions of the NGN in February, 2004: the NGN is a packet network, which provides various services including telecommunication services, and can implement a separation of service functions from bottom-layer transmitting techniques by means of various bandwidths and transmitting techniques with the capability of Quality of Service (QoS); the NGN allows users to freely access networks of different service providers, and supports a universal mobility, which brings consistency and unity in use of the services by users.
A physical model of the NGN which adopts softswitch technique is shown in FIG. 2. The softswitch in FIG. 2 controls MGs through a control protocol such as H.248/Media Gateway Control Protocol (MGCP). The SS7 only runs on the softswitch, and the H.323/Bearer Independent Call Control (BICC)/Session Initiation Protocol (SIP) and other protocols also run on the softswitch while each MG only performs resource operations.
According to the H.248 protocol, the media resource operations on the MG are performed by taking a termination as an operating unit, and termination operations are mainly performed in a context.
The message transmission mechanism of the H.248 protocol is shown in FIG. 3. Those skilled in the art would know that the basic operation unit of an H.248 protocol message is a transaction, which includes several operations to the context and the terminations. A transaction is uniquely identified by a transaction ID (a long word of 32 bits). A combination of several transactions of the softswitch will comprise all the H.248 commands of an entire calling process.
In the MG operations based on H.248 messages, a process of adding terminations T1 and T2 and deleting T2 in a context C1 is shown in FIG. 4. In the figure, the words shown above each arrow illustrates the H.248 message used in each step while the words shown below each arrow gives an explanation of the corresponding H.248 message used in this step.
Detailed description of the H.248 protocol can be found in “Megaco Protocol Version 1.0□RFC3015□” or ITU-T H.248 Specification.
In the above process, once a fault, such as a call setup failure, occurs, the MG must trace the entire process of the call, and output the interacted H.248 messages and the internal resource operations so as to locate and solve the fault. However, since the number of calls and resource operations implemented in an MG is huge, and the timing of these calls and resource operations is interleaved, if information of all the interacted H.248 messages and internal resource operations are outputted, the signaling interaction of the entire process of each call can not be distinguished. Therefore, filters have to be set so that only the signaling interaction of H.248 messages and internal resource operations associated with the call being traced are output.
In the prior art, in order to trace a call on an NGN MG, two technical schemes are often used, wherein both the schemes involve creating a trace task and setting tracing conditions in the MG manually. The implementation of the two technical schemes is as follows:
Technical scheme 1: manually set up a trace task in the MG to trace a designated context. After the reception of an H.248 message by the MG, analyze the context, and judge whether the call that the message belongs to needs to be traced, if the call needs to be traced, output the H.248 messages received and sent and output the internal operation process of the MG associated to the context for use in analyzing the internal processing of the MG.
Technical scheme 2: manually set up a trace task in the MG to trace a designated termination. After the reception of an H.248 message by the MG, analyze the termination in the message, and judge whether the associated call needs to be traced, if the call needs to be traced, output the H.248 messages received and sent and output the internal operation process of the MG associated to the termination for use in analyzing the internal processing of the MG.
In actual applications, both of the two technical schemes have such problems as that the processes of some calls are difficult to trace, the defined tracing conditions are hard to detect and a trace is difficult to implement when there are too many associated contexts or terminations in one call and all the information of which should be outputted.